ObjectivesShift work is associated with myocardial infarction and stroke. We studied if shift work is also associated with incident atrial fibrillation (AF) and if this association differs, depending on sex and age.MethodsWe studied 22 339 participants (age 37.0±9.8 years, 49% women) with paid work from the third (1986–1987), fourth (1994–1995), fifth (2001) and sixth (2007–2008) surveys of the population-based Tromsø Study, Norway. Participants were followed up for ECG-confirmed AF through 2016. Shift work was assessed by questionnaire at each survey. We used unadjusted and multivariable-adjusted Cox regression models to study the association of shift work with 10-year incident AF and incident AF during extensive follow-up up to 31 years. Interactions with sex and age were tested in the multivariable model.ResultsShift work was reported by 21% of participants at the first attended survey. There was an interaction between shift work and age for 10-year incident AF (p=0.069). When adjusted for AF risk factors, shift work was significantly associated with 10-year incident AF in participants <40 years (HR 2.90, 95% CI 1.12 to 7.49) but not≥40 years of age (HR 0.90, 95% CI 0.53 to 1.51). Shift work was not associated with incident AF during extensive follow-up (HR 1.03, 95% CI 0.89 to 1.20). There was no interaction between shift work and sex.ConclusionsShift work was associated with 10-year incident AF in individuals <40 years but not ≥40 years of age. Shift work was not associated with incident AF during extensive follow-up up to 31 years, and there were no sex differences.